Pediatric Nephrology

Pediatr Nephrol (1993) 7:780-784 9 IPNA 1993

Original article Differentiation of intercalated cells in culture G6za Fejes-T6th and Anik6 Nfiray-Fejes-T6th Department of Physiology, Dartmouth Medical School, Borwell Building, Lebanon, NH 03756, USA Received December 3, 1992; received in revised form March 23, 1993; accepted April 26, t993

Abstract. The renal collecting duct is a heterogenous epithelium consisting of intercalated cells (ICCs) and principal cells (PCs). To test the hypothesis that the two cell types might originate from one another and to determine which one of the two is a stem cell, [3-ICCs and PCs were isolated by fluorescence-activated cell sorting and grown on permeable supports. Cultures of sorted PCs maintained their PC phenotype [electrogenic sodium (Na+) reabsorption and potassium (K+) secretion and expression of PC-specific antigens]. In contrast, cultures of sorted ~-ICCs acquired (z-ICC-specific functions (e.g. proton secretion) and gradually expressed functions specific for PCs (amiloride-blockable Na+ current and K + secretion). Most cells in cultures of sorted [3-ICCs also acquired a central cilium, a characteristic feature of PCs. Dual-staining of [3-ICC cultures with cell-specific antibodies against surface antigens revealed that approximately 45% of the cells expressed only ICC-specific antigens and approximately 20% expressed only PC antigens. The remainder of the cells were ICC/PC "hybrids" and stained for both markers. Such hybrid cells were also observed in situ, albeit with a lower frequency, on kidney sections dualstained with cell type-specific markers. The proliferation rate of the two cell types, assessed by pulse labeling cells in S-phase with bromodeoxyuridine or staining with an antibody against a proliferation antigen (KI-67), revealed a significantly higher proliferation rate among [3-ICCs than PCs. In aggregate, these data suggest that [3-ICCs in culture are capable of differentiating into c~-ICCs and PCs and raise the possibility that [3-ICC is the stem cell of the collecting duct. Key words: Intercalated cells - Principal cells - Differentiation - Cell conversion - Ion transport - Antigens Collecting duct

Correspondence to: G. Fejes-T6th

Introduction Distal urinary epithelia contain several cell types intermingled with one another. For instance, in the cortical collecting duct of the mammalian kidney three cell types can be distinguished: principal cells (PCs) and two types of intercalated cells (ICCs). PC, which is the majority cell type is characterized by a polygonal shape, apically positioned nucleus, prominent basolateral infoldings and a central cilium on the apical surface [1]. PCs are engaged in sodium in (Na+), potassium (K +) and water transport and respond to aldosterone and vasopressin [2]. The two types of ICCs (cz- and [3-ICCs) are responsible for the bidirectional transport of bicarbonate, with [3-ICCs secreting bicarbonate and c~-ICCs reabsorbing it [3]. Both types of ICC contain carbonic anhydrase and respond to [3-adrenergic agents [4-6]. At the morphological level, ICCs are distinguished by their high mitochondrial density, a basally positioned nucleus, rounded luminal outline, a dense coat of apical microvilli and microplicae, and the lack of a central cilium [1]. Because of these profound morphological and functional differences, it is generally believed that the various cell types of the collecting duct represent separate lineages. Despite the detailed characterization of the different cell types, it is still not clear how this cellular heterogeneity is established and maintained. To address this question, in the present study we isolated highly purified populations of PCs and [3-ICCs and examined their characteristics in primary culture. These studies revealed that [3-ICCs in culture can differentiate into both o~-ICCs and PCs whereas cultured PCs do not undergo similar transitions.

Methods Cell isolation and culture. PCs and ~-ICCs were isolated from rabbit kidney cortex by fluorescence-activated cell sorting as described previously [6-8]. In brief, small cortical collecting duct (CCD) fragments, containing both PCs and ICCs, were isolated by immunoselection using monoclonal antibody (mAb) ST. 12, which reacts with an ectoantigen on

781

-100

!200

-80

PD 150

-60 > E-40 -20 0

100

-1001

/2~ -50

~ ~-40 SC( 50 < :~ -20 0

20 , 1 I , I -- -50 & 5 6 7 8 A Doysoftenseeding

?pD

'200 5CC'150 '

Mm 15:

[Na+l I[K+] 9 pH

Mm 1,50

j [Na+l I[K+] 9 pH Mm

100 5

0 20 , , i I I ,~-50 3 & 5 5 7 B B Doys oftenseeding

Fig. l. Time course of changes in transepithelial potential difference ( [] ) and short circuit current (@) in monolayers of A cultured principal cells (PCs) and B p-intercalated cells (~-ICCs). Data are means + SE

CCD cells [9, 101. CCD suspensions were then stained with a fluorescein isothiocyanate (FITC)-conjugated PC-specific antibody (DT. 17 [6]) and the {3-ICC marker peanut lectin agglutinin (PNA) [3] conjugated with phycoerythrin, and the two cell types were separated by fluorescence-activated cell sorting [6 - 8]. The purity of sorted [3-ICCs, as assessed by flow cytometric reanalysis and staining with other cell-specific mAbs, was >98%, while that of sorted PC was somewhat lower, approximately 95%. Viability (determined by trypan blue exclusion) of ~-ICCs was 85% • 1% while that of PCs was 68% _+3%. Sorted cells were seeded on porous bottom dishes (Millicell HA, Millipore) at a near saturating density of 4 - 6 x 105 viable cells/cm2 and grown in PC-l-based medium (Ventrex) as described [8-10]. After the cultures reached confluence, medium was changed every 24 h. Transepithelial potential difference (PD) and short circuit current (SCC) were measured in monolayers, using a voltage/current clamp apparatus, as described earlier [10, 11]. Electrical resistance was calculated from the ratio of PD to SCC. Na + and K+ concentrations were determined by flame photometry in media collected after 24 h incubation from the two sides of the monolayers. The pH of the apical and basolateral media was determined following equilibration in 5% carbon dioxide with a pH-sensitive electrode.

50 ;-~-~0 i

50A

i i B

Fig. 2. Basolateral (BL) and apical (A) sodium (Na+) and potassium (K+) ion concentrations in monolayers of cultured PCs (A) and [3-ICCs (B) 6 days after seeding. Data are means +_ SE incubated with a mixture of Texas red-labeled anti-mouse immunoglobulin and FITC-conjugated PNA. Alternatively, the cells were incubated with Texas red-labeled anti-mouse immunoglobulin, and after washing 10% mouse serum was layered onto the coverslips, followed by FITCconjugated DT.17 (PC marker). After a final wash, coverslips were incubated with 1 ~tg/ml 4,6-diamidino-2-phenylindole diluted in 1% paraformaldehyde/0.1% Triton X100/ethanol (3/3/4) to visualize nuclei. Coverslips were mounted in 90% glycerol containing 0.8 mg/ml paraphenylenediamine to retard photo-bleaching, and inspected on a Zeiss Axiophot microscope using a dual FITC/Texas red filter block and an ultraviolet filter block. A similar set of staining was performed by substituting the anti-BrdU antibody with an antibody against KI-67, a nuclear proliferation antigen [15].

Scanning electron microscopy. Cells cultured on Millipore filters were fixed in 3% glutaraldehyde in 0.1 M cacodylate buffer (pH 7.0) for 24 h at room temperature. The specimens were rinsed in the same buffer for 5 rain at room temperature and placed in 0.1 M cacodylate buffer containing 1% osmium and 0.2 M sucrose for 2 h. The samples were dehydrated in a series of ethanol washes, immersed in hexamethyldisilazane for 5 min, air-dried, mounted on aluminum stubs, coated with gold-palladium in a sputter coater and examined under a scanning electron microscope.

Immunocytochemistry. For cytochemistry, cells from confluent monolayers grown on permeable supports were detached and incubated in suspension with cell-specific mAbs, followed by staining with FITC-conjugated anti-mouse immunoglobulinas described previously [9]. Cells were then fixed and mounted in 1% paraformaldehyde and inspected by fluorescence microscopy. On each sample at least 400 cells were counted. Staining for band-3 antigen was performed by the indirect immunoperoxidase technique on acetone-fixed cytospin preparations of cells removed from the monolayers [9] using a mAb (IVF12, generously provided by V. L. Schuster [12]). In a few cases intact monolayers were dual-stained at 0~ with a combination of FITC- and Texas red-labeled cell-specific antibodies for 30 rain, fixed with 1% paraformaldehyde and mounted on coverslips for fluorescence microscopy. Histochemistry was performed on paraffinembedded kidney sections fixed with periodate-lysine-paraformaldehyde as described previously [6, 13]. For the determination of proliferation rates of PCs and [3-ICCs, unfractionated CCD cells, immunoselected with mAb ST.12, were seeded onto glass coverslips at a density of approximately 5 • 10 4 cells/cm2 and at days 3 - 4 they were incubated in an RPMI 1640 medium containing 5% fetal bovine serum, 10 pNI bromodeoxyuridine (BrdU) and 1 gM fluorodeoxyuridine for 4 h. The cultures were then fixed for 10 min in methanol/acetone (95/5) at -20 ~C, air dried and stored at -20~ until staining. The cells were rehydrated in phosphate-buffered saline and incubated with an antibody against BrdU [14], washed and

Results T h e t i m e c o u r s e o f c h a n g e s i n electrical p a r a m e t e r s i n p r i m a r y c u l t u r e s i n i t i a t e d w i t h sorted P C s a n d [3-ICCs is s h o w n in Fig. 1. S o r t e d P C s r e a c h e d c o n f l u e n c e o n d a y 4. A t that t i m e the m o n o l a y e r s d e v e l o p e d a l u m e n - n e g a t i v e v o l t a g e o f 6 m V , w h i c h i n c r e a s e d g r a d u a l l y to a p p r o x i m a t e l y 75 m V b y d a y 7. S C C f o l l o w e d a s i m i l a r t i m e c o u r s e a n d r e a c h e d m a x i m a l v a l u e s o f 50 g A / c m 2 at days 7 - 8 . C h a n g e s i n P D a n d S C C i n c u l t u r e s o f ~ - I C C s foll o w e d a s i m i l a r t i m e course, e x c e p t that these cultures r e a c h e d electrical c o n f l u e n c e 1 d a y earlier a n d started out w i t h a l u m e n - p o s i t i v e v o l t a g e o f a p p r o x i m a t e l y 12 m V a n d a r e v e r s e d S C C o f a p p r o x i m a t e l y 5 g A / c m 2 . T h e transepithelial P D c o n v e r t e d to l u m e n - n e g a t i v e v a l u e s b y day 4 a n d f i n a l l y stabilized at a r o u n d 100 m V w i t h a c o r r e s p o n d i n g S C C o f a r o u n d 150 g P d c m 2. It s h o u l d b e p o i n t e d out that the m a x i m a l S C C a c h i e v e d b y [3-ICC cultures is sign i f i c a n t l y h i g h e r t h a n that o f P C cultures. T h e S C C i n b o t h types o f c u l t u r e s r e p r e s e n t s m a i n l y a N a + c u r r e n t as it c o u l d b e i n h i b i t e d b y the apical a d m i n i s -

782 -30

-3O

PD mV

PD mV O

0 Control *5.SmM +11.6mM KC[ KC[ apicat[y apicaL{y

Control +5.8mM +11.6mM KCt KCL apicatl.y apicat[y

Fig, 3. K + diffusion potentials (PD) in monolayers of cultured PCs (A) and [3-ICCs (B) 6 days after seeding. The medium in control samples contained 5.8 mlVl potassium chloride (KC1).Data are means _+ SE

PC markers

~ o

ST.9

DT.17

~-ICC markers

~- [CC m a r k e r

mH B 63

B 601

Band 3

Fig. 4. Antigenic characteristics of 6-day-old cultures of [3-ICCs. Data are means _+ SE

tration of amiloride. In PC cultures, amiloride decreased PD and SCC to values not significantly different from zero. In contrast, in [3-ICC cultures amiloride unmasked a reversed current of 12.8 + 2.9 gA/cm2. Administration of the carbonic anhydrase inhibitor acetazolamide or acidification of the apical medium abolished this reversed current indicating that it is brought about by electrogenic proton (H +) secretion. In PC cultures these interventions were without effect. Further evidence for electrogenic H§ secretion in [3-ICC cultures and the lack of net H + ion transport in PC cultures was obtained by measuring transepithelial ion gradients across monolayers of the two types of cultures. These results are depicted in Fig. 2. 13-ICC cultures created a transepithelial pH difference of approximately 1.91 + 0.11, whex'eas in PC cultures no significant difference could be detected between the pH values of the apical and basolateral compartments. Also shown in Fig. 2 are the apical

and basolateral concentrations of Na § and K+ after a 24-h incubation period. The development of basolateral-to-apical Na+ gradients provides additional evidence for Na § absorption in both cultures, while the accumulation of K+ at the apical side indicates the existence of active K + secretion. Similar to the amiloride-inhibitable Na § current, Na§ and K+ gradients achieved by cultures of ~-ICCs were significantly higher than those achieved by PC cultures. The mechanism of K + secretion by PCs in the collecting duct involves K+ exit through an apical K+ conductance. The presence of such an apical K+ conductance was confirmed by testing the effect of increasing the concentration of K § in the apical medium on the transepithelial voltage in amiloride-pretreated cultures. These results are summarized in Fig. 3. Stepwise increments in apical K+ concentrations resulted in an increase in PD in both cultures. The slope of the K+-induced increase in PD was 51.4 mV/decade in [~-ICC cultures, suggesting that the apical membrane of cultured [3-ICCs behaves as an almost perfect K § electrode. The slope of this relationship in PC cultures was 25.8 mWdecade. These data suggest that I~-ICCs in primary cultures are capable of acquiring both ot-ICC- and PC-related functions whereas PCs seem to maintain their original phenotype. The pattern of expression of cell-specific antigens is compatible with this interpretation. As can be seen in Fig. 4, in 6-day-old cultures initiated with sorted [3-ICCs, 54%-65% of the cells expressed PC-specific antigens and 25% were positive for band-3 antigen, a marker for ot-ICCs. In addition, 59%-66% of the cells bound ~-ICC-specific markers. Dual staining studies revealed that although hybrid cells with a dual [3-ICC/PC phenotype were also present, on the majority of the cells ~-ICC and PC markers were localized to separate cells. Changes in apical morphology also support the notion that ~-ICCs in primary culture can give rise to PCs. In situ, PCs are characterized by a central cilium, whereas ICCs lack such structures. [3-ICC cultures, fixed 3 days after seeding, at a time when PC-related functions are not yet expressed, exhibited a dense lawn of apical microvilli but lacked cilia (Fig. 5 A). However, on day 8, which corresponds to a time when PC-related functions are fully expressed, the majority of 13-ICC cells had a central cilium (Fig. 5 B).

Fig. 5. Scanning electron micrograph of the apical surface of 3-day-old (A) and 8-day-old (B) cultures of 13-IeCs (magnification x 4,500)

783

BrdU

Ki67

4oi 20J

Ool

PC p-ICC

PC [J-ICC

Fig. 6. Proliferation rate in subpopulations of cuItured collecting duct cells. PCs and [~-ICCs were identified by staining with monoclonal antibody (mAb) DT.17 and peanut lectin agglutinin, respectively. Proliferation rates were determined either by bromodeoxyuridine (BrdU) incorporation (left) or by staining for KI-67 (right). Data are means + SE. Statistical differences (~-ICCs vs. PCs) were calculated using Student's t-test. * P <0.01; ** P <0.001

PNA as ~-ICC marker. The percentage of BrdU-positive cells was significantly higher among those cells that expressed the ~3-ICC marker than among PCs. Similar results were obtained with an antibody against a nuclear antigen (KI67) which is present only in proliferating cells [15]. To explore the in vivo significance of these findings we performed dual labeling studies on renal tissue sections using ICC- and PC-specific mAbs. The rationale behind these experiments was that if interconversion between the different cell types of the CCD also occurs in vivo, one might encounter hybrid cells exhibiting a dual phenotype. Indeed, these experiments revealed the presence of a minority cell population which had dual phenotype. The frequency of ~13-ICC hybrids was 3 % - 5 % as revealed by dual staining with B63 and an antibody against band 3. The percentage of cells exhibiting double staining with an antibody against H+-ATPase (ICC marker [16]) and DT.17 (PC-specific [6]) or with PNA and DT.17 was approximately 1%. Similar observations were made on mouse kidney sections double stained with mAb F13/0121 (PC specific) and an antibody against H+-ATPase (ICC marker, Fig. 7).

Discussion

Fig. 7. Fluorescence images of an initial cortical collecting duct on a 4-gin thick mouse kidney section duN-stained with an antibody against H+-ATPase (red fluorescence, A) and fluorescein isothiocyanate coupled mAb F13/0121 (green fluorescence, B). The two cells on the left stain for H-ATPase only (positive in A, negative in B), while the cell on the right stains for both markers (positive in both A and B). Magnification x 1,012

In aggregate these data indicate that 13-ICCs have the ability to develop into both (x-ICCs and PCs, whereas sorted PCs seem to retain their original phenotype. This raises the possibility that PCs might be terminally differentiated, whereas 13-ICCs might be the stem cells of the CCD. If this were the case, one would expect 13-ICCs to have a higher proliferative capacity than PCs. This possibility was examined by double labeling collecting duct cultures with a cell type-specific antibody and a proliferation marker. The results of four permutations of such staining are summarized in Fig. 6. Proliferating cells were labeled with BrdU, which incorporates into DNA during S-phase, in combination with DT.17, as PC-specific marker or with

The mechanisms that create the cellular heterogeneity of the CCD have not been explored thus far. Our earlier observations [8] and data presented here indicate that [3-ICCs, which in situ are credited with net bicarbonate secretion but not with Na+ and K+ transport, in culture develop into an epithelium in which PC functions, such as electrogenic Na + reabsorption and K + secretion, and o~-ICC functions, such as electrogenic H+ secretion, predominate. These findings, coupled with immunocytochemical data [8] and the morphological observations reported here, point to a remarkable plasticity of these cells and suggest that the typical cellular heterogeneity of the collecting duct probably arises through cellular interconversion. Our observation that single cells of the CCD cell line M-1 [13] can develop into a heterogenous epithelium [8] gives additional support to this conclusion. As there is an inherent difficulty with extrapolating data obtained in vitro to the in vivo situation, further studies are required to establish the in vivo significance of the cellular interconversion observed with cultured cells. Nevertheless, our findings that cells with dual phenotype, like 13-ICC/ o~-ICC hybrids [17] and 13-ICC/PC hybrids [8], are present in both the rabbit and the mouse collecting duct in situ support the assumptions that cellular plasticity also exists in vivo. Questions concerning the stability of the phenotype of ICC subtypes have been raised earlier by Schwartz et al. [18] who provided evidence that [3-ICCs can develop into c~-ICCs following an in vivo acid load. Apparently, conversion of [3-ICCs into c~-ICCs is an unidirectional process, as cMCCs do not develop into ~-ICC even after severe in vivo alkaline loading [19]. Similarly, our finding that [3-ICCs can develop into ~-ICCs and PCs, whereas PCs do not give rise to o~-ICCs or 13ICCs, suggests that ICCs and PCs are not equipotent in their ability to differentiate. As pluripotent cells generally

784

have a higher proliferative capacity, the significantly higher proliferation rate of [}-ICCs than that of PCs, observed in this study suggests that among the collecting duct cells [}-ICCs behave as stem cells and PCs as terminally differentiated ceils. If differentiation of ~-ICCs to PCs is an unidirectional process, we are faced with the obvious question of how can the in vivo observed constant ratio of PCs to ICCs be maintained without eventually depleting the tubule of ~-ICCs. A clue to this problem might come from the peculiar behavior of ~-ICC/PC hybrid cells during division. Based on our preliminary findings [20], such cells can undergo an asymmetric cell division, yielding a bona fide [}-ICC and a PC. This mechanism of asymmetric cell division might provide a means for simultaneous differentiation and stem cell renewal. Acknowledgements. This work was supported by National Institutes of Health grants DK-39523, DK-45647 and DK-41841, and by a Grant-inAid from the American Heart Association (89-985). We also acknowledge the support of the Fanny Rippel Flow Cytometry Facility and the Norris Cotton Cancer Center Core Grant CA 23108.

References 1. gaiz W, Kaissling B (1992) Structural organization of the mammalian kidney. In: Seldin DW, Giebisch G (eds) The kidney, 2nd edn. Raven, New York, pp 707-778 2. Morel F, Doucet A (1992) Functional segmentation of the nephron. In: Seldin DW, Giebisch G (eds) The kidney, 2rid edn. Raven, New York, pp 1049-1086 3. Schuster VL (1990) Bicarbonate reabsorption and secretion in the cortical and outer medullary collecting duct. Semin Nephrol 10: 139-147 4. Brown D, Kumpulainen T (1985) Immunocytochemical localization of carbonic anhydrase on ultrathin frozen sections with protein A-gold. Histochemistry 83: 153-158 5. Manger TM, Koeppen BM (1991) [~-Adreneragic regulation of C1/HCO3 antiporter activity in prima~y cultures of rabbit outer medullary collecting duct cells (abstract). J Am Soc Nephrol 2:706

6. Fejes-T6th G, Nfiray-Fejes-T6th A (1989) Isolated principal and intercalated cells: hormone responsiveness and Na-K-ATPase activity. Am J Physiol 256:F742-F750 7. Fejes-T6th G, NAray-Fejes-T6th A (1991) Fluorescence-activated cell sorting of collecting duct subtypes. J Tissue Cult Methods 13: 173-178 8. Fejes-T6th G, NAray-Fejes-T6th A (1992) Differentiation of renal J}-intercalated cells to c~-intercalated and principal cells in culture. Proc Natl Acad Sci USA 89:5487 -5491 9. Fejes-T6th G, NAray-Fejes-T6th A (1987) Differentiated transport functions in primary cultures of rabbit collecting ducts. Am J Physiol 253:Ft302-F1307 10. N~ay-Fejes-T6th A, Fejes-T6th G (1991) Immunoselection and culture of cortical collecting duct cells. J Tissue Cult Methods 13: 179-184 11. Nfiray-Fejes-T6th A, Fejes-T6th G (1990) Glucocorticoid receptors mediate mineralocorticoid-like effects in cultured collecting duct cells. Am J Physio1259:F672-F678 12. Jennings MJ, Anderson MP, Monaghan R (1986) Monoclonal antibodies against human erythrocyte band 3 protein. J Biol Chem 261: 9002-9010 13. Stoos BA, Nfiray-Fejes-T6th A, Carretero OA, Ito S, Fejes-T6th G (1991) Characterization of a mouse cortical collecting duct cell line. Kidney Int 39:1168 1175 14. GonchoroffNJ, C-reipp PR, Kyle RA, Katzmann JA (1985) A monoclonal antibody reactive with 5-bromo-2-deoxyuridine that does not require DNA denaturation. Cytometry 6:506 -512 15. Gerdes J, Lemke H. Baisch H, Wacker H, Schwab L, Stein H (1984) Cell cycle analysis of cell proliferation associated human nuclear antigen defined by the monoclonal antibody KL67. J Immunol 133: 1710-1715 16. Brown D, Hirsch S, Gluck S (1988) Localization of a proton-pumping ATPase in rat kidney. J Clin Invest 82:2114-2126 17. Schuster VL, Fejes-T6th G, NAray-Fejes-T6th A, Gluck S (1991) Co-localization of H-ATPase and band 3 anion exchanger in rabbit collecting duct intercalated cells. Am J Physio1260:F506-F517 18. Schwartz GJ, Barasch J, A1-Awqati Q (1985) Plasticity of functional epithelial polarity. Nature 318:368 - 371 19. Hayashi M, Schuster VL, Stokes JB (1988) Absence of transepithelial anion exchange by rabbit OMCD: evidence against reversal of cell polarity. Am J Physiol 255:F220-F228 20. Fejes-T6th G, Nfiray-Fejes-T6th A (1991) Celt differentiation and stem cell renewal via asymmetric cell division in the renal collecting duct. JCell Biol 115: 148A

Literature abstract J Med Genet (1989) 26:289-293

A new syndrome of autosomal recessive nephropathy, deafness, and hyperparathyroidism B. D. Edwards, M. A. Patton, S. A. Dilly, and J. B. Eastwood A consanguineous Pakistani family is described in which family members developed renal failure without haematuria, parathyroid hyperplasia, and sensorineural deafness. We believe the condition to be inherited as an

autosomal recessive and to be distinct from Alport's syndrome, which is an X linked condition usually associated with haematuria.